+Intertrace Insights
Prompt injection and exfiltration
Instructions that arrive as data, and the channels that carry your data back out.
- Pieces
- 57
- Threat research
- 16
- Updated
- Oct 20, 2026
6 pieces · newest first
Threat research9 min read
Cross-tool data laundering: when every call is allowed and the export is not
An agent can read restricted data, rewrite it and post it somewhere broader using only permitted tools. Seven beliefs that let this through, what public research and one disclosed case show instead, and the session labels, audience checks and canary tests that catch it.
Intertrace Threat Research
Threat research10 min read
Persistent memory poisoning: the incident that comes back after you close it
Agents that remember can be taught by an attacker once and misbehave for weeks. One fictional case followed from the poisoned write to the investigation, with the published research behind each step and the controls that would have stopped it at the write.
Intertrace Threat Research
Threat research9 min read
Prompt injection beyond text: six myths about images, documents and audio
Vision and audio models read instructions from pixels, PDF layers, screenshots and speech, where text filters never look. Six common beliefs about multimodal injection set against the published research, and the provenance and authority controls that hold when detection misses.
Intertrace Threat Research
Threat research11 min read
Invisible instructions: Unicode tag characters and ASCII smuggling against LLMs
Unicode tag characters mirror ASCII but render as nothing, and models still read and write them. A teardown of the 2024 Microsoft 365 Copilot exfiltration chain, the wider invisible-text family, and a sanitizer for both directions that keeps the evidence.
Intertrace Threat Research
News9 min read
EchoLeak and the exfiltration channel in your assistant's renderer
CVE-2025-32711 let one crafted email pull data out of Microsoft 365 Copilot through an image the client fetched on its own. A step-by-step teardown of the four defenses it got past, and why the fix that transfers is inspecting what leaves.
Intertrace Threat Research
News9 min read
Indirect prompt injection is now an operations problem
Assistants that read email, tickets, web pages and shared documents give every outside author a way to talk to them. Why indirect prompt injection will not be patched away, the strongest objection to that view, and what running it as an operational risk takes.
Intertrace Threat Research
+Stay current
New research, the week it publishes.
Subscribe with RSS, or email support@intertrace.ai with a topic you want covered. To see the controls in action, try the Simulation Lab or read the documentation.